If you're working with cutting-edge vision systems in 2026, this covers the full stack from YOLO26's NMS-free detection to SAM 3's text-guided segmentation and VLM-based reasoning. The standout is how it bridges deployment reality (TensorRT exports, edge optimization, MuSGD training speedups) with modern foundation models. You'll get practical patterns like using YOLO26 for fast proposals then SAM 3 for precise masks, plus the geometry fundamentals (depth estimation, calibration, SLAM) that still matter. The sharp edges table is honest about VRAM and motion blur trade-offs. Best for anyone building real-time vision pipelines where you need both the latest models and actual production knowledge, not just research paper summaries.
npx skills add https://github.com/sickn33/antigravity-awesome-skills --skill computer-vision-expert